AI & ML

dvcrn Bundle

Unified LLM Gateway - One API for 70+ AI models. Route to GPT, Claude, Gemini, Grok and more with a single API key.

dvcrn Bundle

Operate an already-open Hinge session in the browser or on iPhone to review profiles, triage the queue, analyze matches, draft respectful openers or replies, and execute explicit like, reply, or rose actions when the user has asked for autonomous sending. Use when Codex is helping with Hinge browsing, inbox triage, profile analysis, first-message drafting, reply drafting, dating-app note taking, or organizing a live Hinge session. This skill can navigate and read the app, but must not bulk scrape, create fake personas, or impersonate the user.

dvcrn Bundle

Give your AI agent email superpowers via Fastmail JMAP. Read, search, send, move, trash — zero deps. By The Agent Wire (theagentwire.ai)

dvcrn Bundle

Generate and edit images using TensorsLab's AI models. Supports text-to-image, image-to-image generation, plus advanced editing: avatar generation, watermark removal, object erasure, face replacement, and general image editing. Features automatic prompt enhancement, progress tracking, and local file saving. Requires TENSORSLAB_API_KEY environment variable.

dvcrn Skill

Manage Farcaster social interactions, casting, and feed monitoring. Trigger this when users want to post casts, check their feed, or interact with other agents.

dvcrn Bundle

Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.

dvcrn Bundle

Lead Processing Agent

dvcrn Skill

model-failover-doctor

dvcrn Bundle

Monitor, query, and manage New Relic observability data via the newrelic CLI. Covers NRQL queries, APM performance triage, deployment markers, alert management, infrastructure monitoring, and agent diagnostics. Use when user asks about application performance, error rates, slow transactions, deployment tracking, or New Relic configuration.

dvcrn Bundle

DEPRECATED. Use jun-invest-option-master-agent.

dvcrn Bundle

DEPRECATED — duplicate listing. Please use the canonical "openclaw-agent-optimize" skill instead.

dvcrn Bundle

通过 MCP 服务连接 Meego（飞书项目字节内网版），支持 OAuth 认证，可查询和管理工作项、视图等。

dvcrn Bundle

Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use before starting work and after responding to the user.

tinyc0der Skill

Examines agent sessions for harness corrections. Use when asked to reflect on a conversation, conduct a retrospective, or address recurring agent workflow friction.

tinyc0der Bundle

Delegates bounded engineering tasks and phases to Orca agent sessions with configurable runner choices. Use when the user asks for one-shot delegation, to coordinate phases across Codex, Grok, OpenCode, or Antigravity, to supervise a mixed-agent workflow, or to run delegate on a task.

tinyc0der Skill

Discovers and invokes agent skills and selects workflows for engineering requests. Use when starting a session, choosing the skill for an activity, deciding which workflow applies, or routing a scoped change through its lifecycle. Use when classifying work as a feature, bug, epic, or bounded task.

tinyc0der Skill

Optimizes agent context setup. Use when starting a new session, when agent output quality degrades, when switching between tasks, or when you need to configure rules files and context for a project.

tinyc0der Skill

Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensions before it enters the main branch.

tinyc0der Skill

Tests in real browsers via Chrome DevTools MCP. Use when building or debugging anything that runs in a browser. Use when you need to inspect the DOM, capture console errors, analyze network requests, profile performance, or verify visual output with real runtime data. Requires the chrome-devtools MCP server to be configured.

teqfocusadmin Skill

TeqTalk's highest-leverage production skill. Use immediately when a full transcript is available. Produces the 90-second cinematic trailer script for YouTube, LinkedIn, and Reels — using the Diary of a CEO narrative arc model: Problem → Stakes → Surprising Insight → Open Question → CTA. This is NOT a clip or highlight reel — it is a constructed story arc. Also produces: pull quote for text overlay, platform-specific versions (YouTube 90s / LinkedIn 60s / Reel 30s), and the trailer thumbnail brief. This is the single highest-ROI skill in the system. Runs a mandatory 4-iteration refinement loop before Jas approval. Trigger any time a TeqTalk transcript is provided and trailer content is needed.

datnt-cell Bundle

Create and edit C# scripts and ScriptableObjects in Unity, plus scripting-design advisories — assembly definitions (asmdef), script roles, testability, patterns, and async-model choice. Use when the user wants scripts written/read/updated through Editor automation, or wants review/guidance on script architecture. Read unity-core first for the REST protocol before calling any skill named here.

tips0 Skill

Generate robust AI skills via iterative interview

nvidia Bundle

Run srt-slurm (srtctl) inference benchmarks on a Slurm cluster. Use when setting up a checkout, writing or migrating a recipe, submitting a job, or reading its results.

jrudman25 Skill

Research, audit, compare, and recommend installable agent skills when the user explicitly asks to find a skill, recommend skills, compare skill options, extend agent capabilities, or decide whether a recurring workflow should become a skill. Do not invoke merely because an ordinary task could theoretically be handled by an installable skill.

qingmou314-design Skill

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

chenycl Bundle

Browser use / computer use where TypeSafe's Jev (System One model) supplies calibrated judgments and code owns the control loop, in the style of NanoJev's "atomic judgments + code planning". Use when an agent must accomplish a goal in a real browser (open pages, search, sign in, fill forms, add to cart, start a trial, reveal content) without a vision model: it drives ego lite by default (reusing the user's logged-in Chrome sessions), or Chrome via DevTools protocol, or Safari via safaridriver. Also use for any raw Jev judgment (noul / choice / score) from the CLI or MCP. Triggers: "browse", "open the site and …", "use the browser to …", "jev browser", "computer use", "自动操作浏览器", "用浏览器完成".

minthcm Bundle

Use this skill whenever using the MintHCM MCP to check whether the task you are about to perform, or the information you are looking for, is already in the documentation.

andylumax33-glitch Bundle

Convert a research paper and its publicly available materials into a traceable, resource-first MCP agent package.

pavelzw Bundle

Guidelines for evaluating non-deterministic functions with pydantic-evals. Use when writing evals, defining datasets and cases, creating custom evaluators, or testing AI agent outputs with pydantic-evals.

cheshiremew Bundle

用于判断一项能力应由模型或 Agent、确定性程序、软件或服务还是组合完成，并选择提示词、模板、Skill、CLI、API、GUI 或插件等合适包装与入口；也用于在创建前查找已有方案，完整理解、创建、改造、审查、蒸馏、迁移、评估和分发 Codex/Agent Skill。用户想选择能力载体，判断某个 Skill 是否好用、为什么走偏、职责是否过宽或过窄、该保留或精简什么，想从对话、项目或另一 Skill 中提炼并吸收可复用能力，设计自我进化、运行资源或完整流程图，或者检查入口、边界、界面信息和真实使用结果时使用。

ycr97 Bundle

Teach or learn a complex concept deeply by extracting its governing principle, building a mental model, explaining the real mechanism, testing misconceptions, requiring active application, using Socratic correction, and verifying transfer. Use when the user asks to learn, understand, master, or deeply explain a topic rather than receive only a direct factual answer. Do not use for simple lookups or when the user explicitly wants only a concise answer.

benjaminstelzer Bundle

Ask Claude Code and a fresh normal Codex Astra project task in parallel for independent read-only opinions, reviews, critiques, comparisons, or alternative analysis. Use when the user explicitly asks for both Claude or Fable and Astra, says "Frage Fable und Astra", or invokes $ask-claude-and-astra-for-codex. Do not use this Skill instead of ask-astra-for-review-for-codex, ask-sol-for-review-for-codex, ask-claude-and-sol-for-codex, ask-claude-for-codex, or for a generic request for another model or two opinions.

huketo Skill

Ask the human a question and block until they answer, in the agent's interface or over Telegram or Discord. Use when a decision, approval, credential, or missing requirement needs a human who may be away from the terminal, such as a destructive or irreversible action, a real design tradeoff, a value only the human has, a contradictory requirement, or the end of a long unattended run.

cskwork Bundle

Build and iteratively improve agent skills with evaluations, benchmarks, and trigger optimization. Use when a user wants to test or refine a new skill, modify an existing skill, compare skill versions, measure performance, or optimize a skill description. For a simple first-draft skill with no evaluation or iteration, use write-a-skill instead.

boshu2 Bundle

Run one prompt through headless Codex and capture its result. Use when: requesting a single noninteractive Codex process. Not for worker batches or retries.

boshu2 Bundle

Dispatch independent tasks to parallel workers or selected persistent roles. Use when: delegation is authorized with disjoint scopes; execution does not validate output.